Fig. 2From: Surrogate endpoint evaluation using data from one large global randomized controlled trialCorrelation between three-component MACE (surrogate endpoint) and all-cause death (true endpoint). a Analysis of the LEADER dataset grouped while ensuring ≥ 30 deaths (any cause) in each resulting group; 14 groups were derived. b Secondary analysis of the dataset grouped while ensuring ≥ 20 deaths in each resulting group; 17 groups were derived. c Secondary analysis of the dataset grouped while ensuring ≥ 40 deaths in each resulting group; 11 groups were derived. Circles represents a group by country in LEADER; the size of the circle is proportional to the number of trial participants in the group. The coefficient of determination (R2) and the associated 95% CI were derived from a logistic regression model of the treatment effect (HR between liraglutide and placebo) for the surrogate endpoint (MACE) vs that for the true endpoint (all-cause death). CI confidence interval, HR hazard ratio, MACE major adverse cardiovascular eventBack to article page